Social and structural HIV prevention in alcohol-serving establishments: review of international interventions across populations.
Article Type: Report
Subject: Bars, saloons, etc. (Health aspects)
Bars, saloons, etc. (Research)
Drinking of alcoholic beverages (Influence)
Drinking of alcoholic beverages (Physiological aspects)
HIV infection (Risk factors)
HIV infection (Development and progression)
HIV infection (Research)
Sex (Research)
Sex (Health aspects)
Disease transmission (Research)
Sexually transmitted diseases (Prevention)
Sexually transmitted diseases (Research)
Author: Kalichman, Seth C.
Pub Date: 09/22/2010
Publication: Name: Alcohol Research & Health Publisher: U.S. Government Printing Office Audience: Academic; Professional Format: Magazine/Journal Subject: Health Copyright: COPYRIGHT 2010 U.S. Government Printing Office ISSN: 1535-7414
Issue: Date: Fall, 2010 Source Volume: 33 Source Issue: 3
Topic: Event Code: 310 Science & research Canadian Subject Form: Sexual behaviour; Sexual behaviour
Product: Product Code: 8000147 Venereal Disease Prevention; 5813000 Drinking Places NAICS Code: 621999 All Other Miscellaneous Ambulatory Health Care Services; 72241 Drinking Places (Alcoholic Beverages) SIC Code: 5813 Drinking places
Geographic: Geographic Scope: South Africa; Zimbabwe Geographic Code: 6SOUT South Africa; 6ZIMB Zimbabwe
Accession Number: 243042490
Full Text: Alcohol use is among the most reliable predictors of sexual risk behavior across populations at high risk for human immunodeficiency virus (HIV)/acquired immune deficiency syndrome (AIDS). Associations between alcohol use and sexual risks are observed regardless of whether alcohol use is measured in terms of global patterns of drinking, or within situations where drinking occurs, or at the level of sexual events (Weinhardt and Carey 2001). Alcohol elevates sexual risks through multiple channels, including the psychological effects of intoxication on decisionmaking, inhibition of protective actions, interactions between drinking and risk taking personality characteristics, and perceived expectations regarding the effects of alcohol use on risk-taking and sexual experiences (Cook and Clark 2005). The association between drinking and sexual risk behaviors has led to interventions that seek to reduce both alcohol use and sexually risky behaviors (Palepu et al. 2005). For example, an HIV risk reduction intervention for inner-city African-American men reduced alcohol use in sexual contexts by 85 percent from baseline (Kalichman et al. 1999). In a similar study of one-on-one risk reduction counseling in South Africa, researchers reported a 59 percent reduction in alcohol use in sexual contexts 3 months following the intervention (Kalichman et al. 2007). However, the intervention effects on sex-related alcohol use decreased within 6 months to a 43 percent reduction from baseline. This short-lived effect is likely the result of individuals returning to unchanged high-risk environments that support risk-related alcohol use (Kincaid 2004), suggesting the need for interventions to alter risk factors within alcohol-serving environments.

This article examines the alcohol-serving venues that intersect with HIV transmission risks and offer potential environments for HIV prevention. Venues of greatest interest are those that sell alcohol for onsite consumption and serve as a point of social interaction. Multiple types of drinking establishments fit these criteria across cultures, including neighborhood bars, dance clubs, karaoke bars, taverns, wine shops, and informal lounges. In addition to describing the intersection of drinking places with HIV/AIDS risks, this article reviews the literature on HIV prevention interventions based in alcohol venues. Venue-based interventions are defined as strategies designed to alter the social and/or physical environment to promote protective action and reduce risk. Therefore, this article does not include interventions that target drinkers or those that are delivered outside of drinking places. The outcomes from completed venue-based intervention trials point to gaps in the existing literature and suggest future directions for HIV prevention efforts based in alcohol-serving venues.


The contextual factors within which alcohol and sexual behaviors intersect are critical to our understanding of how alcohol influences HIV risks. The environments of drinking establishments are multifaceted and serve multiple functions, including being a place for recreation, fostering socially sanctioned drinking, facilitating social relationships, and providing opportunities to meet sex partners. The subsequent sections review selected studies that describe sexual risks for HIV incurred within drinking places. These studies are summarized in Table 1.

Studies of Gay Men

Early studies of HIV risks in alcohol serving establishments were focused on gay bars, primarily in the United States and Europe. Across studies, approximately one in three men who have sex with men sampled from gay bars and clubs reported engaging in unprotected anal intercourse, the highest risk behavior for HIV transmission. Nardone and colleagues (2001), for example, found that one-third of men who go to gay bars in London and Edinburgh engaged in unprotected anal sex, and only half of the men had been tested for HIV A study in California showed that one in three men who visit gay clubs and bars reported five or more sex partners in the past year (Xia et al. 2006). Men who socialized at gay bars also demonstrated higher risk for HIV infection than gay men who did not go to bars, although men who drank at bars were at lower risk than men who attended sex-focused "circuit" parties. Another study showed that men who meet sex partners at gay bars may not differ significantly from men who meet partners at other open gay venues, such as bathhouses, private sex parties, gyms, and the Internet (Grov et al. 2007). Although behaviors across gay community venues may not differ, the prevalence of disease does vary. In China, for example, HIV (0.8 percent) and syphilis (2.4 percent) prevalence in gay bars is substantially lower than in gay saunas (3.6 percent and 20.7 percent, respectively) (Hong et al. 2009). From an HIV prevention perspective, gay bars may therefore be primarily social venues where sex partners meet and drinking may merely be a contextual rather than causal factor.

Other High-Risk Populations

Commercial sex workers represent a risk population that also is frequently accessed in drinking places. Female sex workers in Malawi, for example, who drink and contract a sexually transmitted infection (STI) are less likely to use condoms than their counterparts who also contract STIs but do not drink (Zachariah et al. 2003). Women who socialize in drinking places, whether involved in sex work or not, demonstrate a high risk for HIV infection (Kapiga et al. 2002; Singh et al. 2009). Meeting sex partners in drinking places increases the likelihood of sex after drinking and reduces condom use (Mataure et al. 2002; Sivaram et al. 2008).

Analyzing the Overlap of Drinking Venues and Sex-Partner Meeting Places

Concurrency of sex partners within and outside of drinking places is a significant facet of HIV/AIDS in southern Africa (Carter et al. 2007; Mah and Halperin 2008) and is facilitated by the sexual networks within drinking establishments. Sex partners often meet at informal alcohol-serving establishments that are common throughout southern Africa, such as private homes where alcoholic beverages are sold and served (Morojele et al. 2006; Weir et al. 2003). For example, one study (Kalichman et al. 2008) that surveyed men and women recruited from informal local township bars (e.g., shebeens) in South Africa found that more than one in four drinkers reported having met sex partners at their local shebeen. People who meet sex partners in drinking venues report greater numbers of recent sex partners and higher rates of unprotected intercourse compared with people who did not meet partners in drinking places. Only half of those surveyed had been tested for HIV in a community in South Africa where one in five people are HIV infected (Shisana et al. 2005). These patterns of risk create urgency for HIV prevention in drinking venues.

Using a well-developed system for rapid community assessments, Weir and colleagues (2003) mapped the linkages among places where people meet new sex partners and places where people drink alcohol. Studies that have used Weir's PLACE methodology show a consistent overlap among drinking behaviors and sexual risks in alcohol-serving venues (Figueroa et al. 2007; Khan et al. 2008). For example, in Zimbabwe, 47 percent of women ages 15-19 years surveyed at night clubs and other drinking places had two or more sex partners in a 1-year period compared with less than 5 percent of young women surveyed outside of drinking places (Singh et al. 2009). In South Africa, more than 85 percent of the places where people meet sex partners are alcohol-serving venues (Weir et al. 2003).

The overlap between drinking places and venues where people meet sex partners was observed in both urban and rural areas. Across three cities in South Africa, between 78 percent and 87 percent of new sex partners met at shebeens. As many as half of the people who drink at shebeens report having had two or more sex partners in the previous 2 weeks, and more than half of men and one in five women who have met partners in shebeens have had sex on the premises (Kalichman et al. 2008). Not surprisingly, informal alcohol-serving establishments rarely display HIV prevention messages or have condoms available for their customers (Figueroa et al. 2007; Weir et al. 2003).

Homes Where Alcohol Is Sold

Alcohol also is often sold from lounges in people's homes that double as small alcohol-serving businesses. People who live in alcohol-serving home businesses are at increased risk for HIV/AIDS. In Ugandan villages, for example, only 4 percent of people live in homes that serve alcohol, but the HIV prevalence in these homes is 15 percent, nearly double that of the surrounding community (Mbulaiteye et al. 2000).

Beer Halls

HIV risks also are observed in larger drinking venues, such as taverns and beer halls. Basset and colleagues (1996) reported that the HIV prevalence in Zimbabwean beer halls (i.e., large social venues that primarily serve men) is as much as twice that of the general Zimbabwe population. Sexual mixing patterns within beer halls facilitate HIV transmission, especially when other STIs are prevalent. For example, 60 percent of men and 40 percent of women who have multiple sex partners in Zimbabwe drink at beer halls (Lewis et al. 2005). Fritz and colleagues (2002) found similar behaviors in these venues and showed that HIV prevalence increases with greater frequency of drinking at beer halls.

Drinking Venue Employees

Employees as well as patrons of alcohol serving establishments are at higher risk for HIV infection. For example, men who work in Tanzanian bars and hotels who drink at least once a week are significantly more likely to have herpes simplex virus infection, a known marker for HIV transmission risk, than other men surveyed at the same places who do not drink (Kapiga et al. 2002, 2003). Women working in food and recreational businesses near southern-African gold mines who drink are significantly more likely to have HIV infection than other women who drink in the communities that surround the mines but do not work in drinking venues (Clift et al. 2003).

Social and Sexual Networks

Higher rates of unprotected sex and STIs in contexts with high HIV prevalence create highly favorable circumstances for HIV transmission. HIV can rapidly spread within drinking venues that have tightly knit social and sexual networks which foster overlapping sexual mixing patterns. Alcohol- serving establishments amplify HIV transmission risks by providing a place where sex encounters occur with rapid turnover of partners (Fritz et al. 2002). Alcohol establishments are often themselves sex venues, where back rooms, back corners, and adjacent buildings or shacks offer locations for sex (Morojele et al. 2006). Beyond the social interactions that naturally occur in drinking places, the use of alcohol itself and its association with sexual risk behaviors makes these environments compelling venues for HIV risk reduction. The social and physical structures of alcohol drinking places therefore create opportunities for community-based HIV prevention interventions.


A review of the literature on HIV interventions delivered within alcohol-serving establishments yielded studies including randomized trials, quasi-experimental studies (i.e., interventions in which the intervention groups and control groups were not randomized), and public health evaluations. The interventions evaluated in these studies (summarized in Table 2) targeted diverse populations in the United States, the United Kingdom, Asia, and Africa and concentrated on altering the social and/or physical environments of drinking places to reduce HIV transmission risks. For the most part, the target populations were consistent with the HIV risk populations of the countries. Specifically, gay bars were targeted in the United States and United Kingdom, whereas female sex workers were the focus in Asia, and in Africa the target has been drinking places that serve local communities within generalized HIV epidemics. The interventions are characterized by their (1) targeting social interactions, (2) changing structural and environmental features, and (3) targeting both social and structural levels.

Social Influence Interventions

The majority of alcohol venue-based interventions target social environments to influence behavior. Such interventions aim to change the perception that a particular behavior is socially acceptable (i.e., alter the perceived social norms) and influence the interactions of people in the environment to shape and sustain protective social norms.

Of nine trials identified, five focused on social influences for HIV prevention within drinking venues. Four interventions explicitly tested the popular opinion leader (POL) model (Kelly et al. 1997), and one evaluated a peer-led intervention that approximated the central features of the POL model. The POL model identifies, trains, and motivates key community-defined opinion leaders to act as agents of change by shifting social norms and thereby behaviors toward safer practices. The POL intervention is grounded in diffusion of innovation theory (Rogers 2003), which specifies that trends and innovations are initiated by a relatively small segment of a population that acts as natural opinion leaders within environmentally or geographically demarcated populations. Innovations that are adopted, modeled, and endorsed by opinion leaders diffuse across social gradients throughout a population. Opinion leaders are trained in interpersonal communication skills to enhance their sense of self-efficacy for acting as agents of change. Although generalized to other settings, the POL model was designed for use in drinking places in small cities, where bars may offer the only outlet for gay men to socialize.

The POL intervention requires a small cadre of trusted, well-liked people who frequent the bar to be trained to initiate conversations and endorse safer sex practices. The active ingredient in the intervention is therefore informal casual conversations with peers, rather than formal education. Using conversations in natural settings as the basis for delivering the intervention is the most compelling feature of the POL model. In these conversations, POLs can correct misinformation; discuss the importance of HIV in the community; talk with network members about strategies to reduce risk, such as keeping condoms nearby; avoiding sex when intoxicated; and resisting unsafe sex. The intervention uses environmental prompts to serve as conversation starters, such as posters or t-shirts with messages and logos. The intervention core elements are (1) identifying and enlisting the support of popular and well-liked opinion leaders to take on risk-reduction advocacy roles, (2) training cadres of opinion leaders to disseminate risk-reduction endorsement messages within their own social networks, and (3) supporting and reinforcing successive waves of opinion leaders to help shape social norms to encourage safer sex (Kelly 2004).

POL Studies Set in Small U.S. Cities. Kelly and colleagues (1992) reported significant reductions in high-risk sexual practices in smaller cities in the southern United States resulting from the POL intervention. The study used a quasi-experimental design with multiple baselines and staggered intervention implementation. The first city to receive the POL intervention had two baselines and three 3-month (9 months)-interval postintervention implementation followups. Two additional cities implemented the intervention after four baselines, with one city having two (6 months) and the other having one (3 months) postintervention followup. The pattern of results was consistent across the three cities: the POL intervention demonstrated reductions in men reporting multiple sex partners, unprotected insertive and receptive anal sex, and increased condom use. The magnitude of change was significant, with up to 29 percent reductions from baseline levels. These behavior changes were bolstered by theoretically predicted changes in perceived social norms supportive of safer sex.

Following the quasi-experimental study, Kelly and colleagues (1997) conducted an experimental test of the community-level POL intervention with eight matched pairs of smaller U.S. cities randomized to either receive the POL program or a comparison intervention that placed general HIV prevention education materials and condoms in the bars. The intervention implementation in half of the pairs of cities was systematically confounded with changes in season, possibly playing havoc with the intervention implementation and assessments. The outcomes were therefore based on the remaining four pairs of cities. The findings mostly replicate the behavior changes observed in the earlier quasi-experimental study, demonstrating 30 percent reductions in unprotected anal intercourse and 50 percent increases in condom use.

POL Study Set in New York City. A study that implemented the POL model in New York City successfully replicated Kelly and colleagues' (1992) quasi-experimental trial (Miller et al. 1998, 2003) among an extremely high-risk group of male sex workers (i.e., hustlers). The adapted intervention demonstrated significant reductions in paid sex encounters and unprotected anal sex in paid sexual relations. In addition, moderator analyses showed that White and Latino men reduced their frequency of unpaid unprotected anal intercourse in response to the intervention, with no such reduction observed among Black men. This pattern of results was similar to that observed by Kelly and colleagues (1992), although with a somewhat smaller effect size.

POL With Community Interventions. As part of a larger multicountry trial (National Institute of Mental Health [NIMH] 2008), researchers adapted the POL intervention to include additional elements of community mobilization derived from non-bar-based community-level HIV prevention interventions (Kegeles et al. 1996). The adapted version was called the community-POL intervention, or C-POL. Researchers conducted the C-POL intervention in five countries, and in one country the implementation occurred in alcohol-serving establishments. Specifically, the intervention was delivered at wine shops in slums of Chennai, India. Wine shops serve mostly beer and distilled alcohol as well as light food and have bars attached to them where patrons can sit and socialize. Such shops primarily serve men, although women frequently meet men outside or near the wine shops. A cluster of wine shops served as the defined intervention venue. The intervention group demonstrated substantial reductions in unprotected sex behaviors and reductions in STIs over the 2-year observation period (NIMH Collaborative HIV/STD Prevention Trial 2008). However, similar reductions in risk and disease were observed in wine shops that served as controls, where only nontailored prevention messages were delivered. Specifically, wine shops exposed to the C-POL intervention had an STI incidence of 7 percent compared with 8 percent in the nonintervention wine shops. Therefore, results of the C-POL intervention on wine shop patrons in India did not demonstrate significant effects on HIV transmission risk behaviors or on biologically assessed STIs.

Peer-Led Intervention. Fritz and colleagues (2009) implemented a social influence intervention that targeted men who socialize and drink in Zimbabwean beer halls. The intervention used a peer-led education model that emphasized men talking with each other about HIV/AIDS prevention. Men volunteered to attend a 3-day training to learn about HIV and how they can help their friends avoid HIV infection. The training included elements of information/education, motivational enhancement, and behavioral skills training for HIV risk reduction. Men who became "peer educators" agreed to volunteer and deliver prevention messages for 15 months posttraining. Like the POL/C-POL model, this intervention was aimed at diffusing prevention strategies and messages throughout the beer hall to change social norms and behavior. Twelve beer halls were randomly assigned to receive either the peer influence intervention or a standard message and access to condoms. The intervention failed to demonstrate any significant effects on behavioral risk indicators, including unprotected sex acts within and across specific partner types, condom use, and numbers of partners. For example, men in the beer hall intervention sites reduced their unprotected sex by 11.8 episodes compared with a reduction of 15 episodes among men in the control venues. The reductions were both significant but not significantly different from each other. Similarly, men in the intervention beer halls reported an average of five episodes of unprotected intercourse with nonwife partners, as did the men in the control beer halls.

Structural Interventions

Interventions that aim to influence behavior through changes in law, policy, and the physical environment generally are referred to as structural interventions (Blankenship et al. 2000). In HIV prevention, examples of structural interventions include mandatory HIV testing for sex workers, closing of public sex environments, and legal needle and syringe exchange programs. Structural interventions also can include improved services that are meant to avert HIV transmission, such as enhanced STI detection and treatment and improved HIV prevention services (Bauermeister et al. 2008). Only two studies have reported on the use of structural approaches to HIV prevention in alcohol-serving establishments. Both studies were conducted as public health program evaluations that did not include comparison conditions. Rou and colleagues (2007) implemented strategies to enhance HIV/STI education and diagnostic and treatment services for female sex workers in bars, massage parlors, dance halls, and beauty parlors in two provinces in China. The intervention included active outreach to women in the venues as well as condom promotion campaigns. Over 1 year of observation, the program showed significant improvements in HIV prevention knowledge, reductions in unprotected sex, and increased condom use among women across venues. Prevalence of incident STIs dropped dramatically during the study period.

In a similar approach taken with men who have sex with men in New York City, Blank and colleagues (2005) reported the effects of the "hot shot" public health program. Aimed at men in gay bars, the intervention offered STI screening and diagnostic and treatment services in alcohol venues. The program reached over 1,600 men in six clubs and three community-based organizations and delivered health services to between 12 percent and 44 percent of event attendees. The program was deemed successful in terms of its implementation and detection of new cases of STIs and HIV. However, changes in behavior or in disease outcomes were not reported.

Combining Social Influence and Structural Interventions. A potentially powerful approach to HIV prevention ' in alcohol-serving establishments is the combination of both social influence and structural interventions. These multilevel models attempt to simultaneously shift social norms to reinforce protective behaviors as well as enhance services and change policies toward HIV prevention. Two studies reported the effects of multilevel social and structural interventions delivered in drinking places. Flowers and colleagues (2002) tested the Gay Men's Task Force intervention in Glasgow, Scotland. This intervention had three major components, one social influence and two structural, each of which had been shown effective in the past and combined for the first time. The intervention included peer-led sexual health promotion conducted in five gay bars by 42 peer educators over a 9-month period. Peer educators, modeled after opinion leaders in the POL intervention, received 2 days of training focused on communication skills for promoting safer sex messages. A second component of the intervention established enhanced STI and sexual health services for gay men as well as a free sexual health hotline service. Results from the two-city quasi-experimental trial demonstrated a significant effect for engaging men in STI services, as evidenced by increased hepatitis B vaccination and HIV testing. Among men who had conversations with peer educators, 49 percent had contemplated and 26 percent reported making changes in their sexual behaviors. However, there was no evidence of community-wide sexual behavior changes resulting from the intervention (Hart et al. 2004).

The second multilevel intervention targeted women sex workers in bars, discos, and night clubs in the Philippines (Morisky et al. 2006). The social influence intervention component was grounded in peer counseling. Female sex workers were enlisted and trained in HIV risk education and prevention actions, focusing on condom use and sexual negotiation skills. The structural component of the intervention enlisted the managers of the drinking places to attend HIV education sessions. The managers were trained to implement a continuum of HIV prevention policies and practices. The training focused on promoting employee health, with specific emphasis on the diagnosis and treatment of STIs. The quasi-experimental study used cities as the unit of randomization, with four cities randomized to four conditions. One site received the peer-counseling intervention, one site received the structural manager intervention, a third site received both the peer-influence and manager components (multilevel condition), and the fourth condition received neither component. Results showed positive effects for both the individual peer counseling and manager components. However, the combination of peer counseling and manager training demonstrated the greatest impact on condom attitudes, venue policies favoring condom use, and reductions in subsequent STIs over the 2-year observation period.

Summary of Intervention Findings

Alcohol venue-based HIV prevention interventions demonstrate a mixed pattern of results. Social influence interventions, particularly the POL model, appear effective in U.S. gay bars, including bars in small and large cities. What these venues have in common are close-knit social networks within which community members know and often trust each other. Gay communities have few public places ' where people can meet and socialize. Although the social landscape is changing in gay communities (Simon Rosser et al. 2008), gay bars historically have been the focal point. Still, social influence models closely tied to the POL model did not work in U.K. gay bars, Indian wine shops, and Zimbabwean beer halls. Understanding the characteristics of social networks and environments that account for the effects of social influence interventions will determine the ultimate utility of these approaches.

The structural interventions reviewed here demonstrated promising outcomes. Enhancing STI and sexual health services through venue-based programming increased hepatitis vaccination and HIV testing in the U.K. and reduced STIs in women involved in sex work in China and men attending gay bars in New York City. Most promising are the additive effects of social influence and manager training to improve sex-worker health in the Philippines. The pattern of results from these few trials suggest that interventions that simultaneously target individual, social, and structural risk factors may hold the greatest promise for alcohol venue-based HIV prevention.


Models of HIV risk reduction focused on the individual may be insufficient for reducing alcohol-related HIV risks because they do not address social, structural/environmental, and contextual influences on behavior. Individual level models rarely achieve sustained behavior change due to unaltered risk environments, the potency of peer pressure, social reinforcements for maintaining risk behaviors, and the interaction between alcohol use and alcohol environments in promoting sexual risks. Interventions that address social influences on behavior, such as social norms and interpersonal contingencies of reinforcement (such as in the POL model), will help support initial behavior changes for potentially long durations. In addition, changes to the physical environment, including policies that impact behavior, also can maintain risk-reducing practices. Interventions that integrate individual, interpersonal, and environmental levels of change will likely prove most effective in reducing alcohol-related HIV risks.

Individual-level intervention components aim to influence motivation and build problem-solving skills, condom use, and sexual negotiation skills. Individual skills-based HIV prevention interventions have proven effective at reducing HIV risk behaviors among drinkers, albeit for short periods of time (Kalichman et al. 2007). Individual skills-building interventions also can include communication skills training geared toward influencing social interactions and social networks within drinking environments. Finally, environmental and policy interventions can support the individual and social influence changes by creating more protective environments. Although there are few structural interventions for sexual risk behavior, policies that regulate drinking age, maintain safety standards of drinking environments, offer health services to sex workers, and provide universal access to STI treatment will likely have a significant impact on HIV risks.

Multilevel alcohol venue-based interventions are feasible for implementation in alcohol-serving establishments. Weir and colleagues (2008) describe environmental components that fit well within the parameters of multilevel interventions. Social cues and prompts can be placed in drinking environments by using posters to communicate prevention messages. Condoms also can be made accessible in venues using dispensers and condom machines with signage. Weir and colleagues (2008) suggest using outreach workers with brief surveys to help engage patrons and raise awareness. Finally, events that support HIV prevention, such as drama and music events, can be implemented in drinking places and include onsite HIV testing and STI screening. Structural changes in the drinking environment will both foster and sustain behavior change that occurs with intensive interventions delivered at individual and social levels. As suggested by Morisky and colleagues (2006), structural changes can be implemented gradually and in synch with the current environment and policies.

HIV prevention behaviors can be adopted, accepted, and become normative within social networks (Kelly et al. 2006; Latkin et al. 2004), and these processes can occur within drinking environments. A subset of influential community members can encourage behavior change through informal social interactions in existing relationships within drinking places. Concepts derived from network-based interventions can be applied though multilevel approaches and provide a mechanism for altering social environments to shape, support, and reinforce behavior change. Interventions can be designed to use existing social networks in drinking places. Peer-driven interventions, for example, that have been successful in U.S. gay bars and among injection drug using populations can be enhanced by having community health and outreach workers deliver prevention messages and resources, such as condoms. Drinking environments can provide venues for chain recruitment of successive waves of peers. At the structural level, drinking environments can post prevention messages that encourage conversation. Venue owners can be enlisted to help shape the house rules, enforce drinking laws, provide condoms, invite health department screenings and onsite testing, and promote the health of their staff as well as clients.

Historically, HIV prevention science has examined intervention components one level at a time. Results from single-level interventions have been discouraging, whereas interventions implemented at multiple levels have been more encouraging. Thus, it would appear that the most promising approaches to effective alcohol venue-based interventions are those that are multilevel and sustained over long periods of time. Multilevel HIV prevention interventions in drinking venues therefore warrant further investigation.


The author declares that he has no competing financial interests.


BASSETT, M.T.; MCFARLAND, W.; RAY, S.; ET AL. Risk factors for HIV infection at enrollment in an urban male factory cohort in Harare, Zimbabwe. Journal of Acquired Immune Deficiency Syndromes and Human Retrovirology 13:287-293, 1996. PMID: 8898675

BAUERMEISTER, J.A.; TROSS, S.; AND EHRHARDT, A.A. A review of HIV/AIDS system-level interventions. AIDS and Behavior 13:430-448, 2008. PMID: 18369722

BLANK, S.; GALLAGHER, K.; WASHBURN, K.; AND ROGERS, M. Reaching out to boys at bars: Utilizing community partnerships to employ a wellness strategy for syphilis control among men who have sex with men in New York City. Sexually Transmitted Diseases 32(Suppl. 10):S65-S72, 2005. PMID: 16205296

BLANKENSHIP, K.M.; BRAY, S.J.; AND MERSON, M.H. Structural interventions in public health. AIDS 14(Suppl. 1):S11-S21, 2000. PMID: 10981470

CARTER, M.W.; KRAFT, J.M.; KOPPENHAVER, T.; ET AL. "A bull cannot be contained in a single kraal:" Concurrent sexual partnerships in Botswana. AIDS and Behavior 11:822-830, 2007. PMID: 17295072

CLIFT, S.; ANEMONA, A.; WATSON-JONES, D.; ET AL. Variations of HIV and STI prevalences within communities neighbouring new goldmines in Tanzania: Importance for intervention design. Sexually Transmitted Infections 79:307-312, 2003. PMID: 12902582

COOK, R.L., AND CLARK, D.B. Is there an association between alcohol consumption and sexually transmitted diseases? A systematic review. Sexually Transmitted Diseases 32:156-164, 2005. PMID: 15729152

FIGUEROA, J.P.; DOLAN, C.B.; DALE, D.; ET AL. An assessment of sites where persons go to meet sexual partners in St. James, Jamaica, using the PLACE Method. Sexually Transmitted Diseases 34:410-415, 2007. PMID: 17038963

FLOWERS, P.; HART, G.J.; WILLIAMSON, L.; ET AL. Does bar-based, peer-led sexual health promotion have a community-level effect amongst gay men in Scotland? International Journal of STD and AIDS 13:102-108,2002. PMID: 11839164

FRITZ, K.E.; WOELK, G.B.; BASSETT, M.T.; ET AL. The association between alcohol use, sexual risk behavior, and HIV infection among men attending beer halls in Harare, Zimbabwe. AIDS and Behavior 6:221-228, 2002.

GROV, C.; PARSONS, J.T.; AND BIMBI, D.S. Sexual risk behavior and venues for meeting sex partners: An intercept survey of gay and bisexual men in LA and NYC. AIDS and Behavior 11:915-926, 2007. PMID: 17206536

HART, G.J.; WILLIAMSON, L.; AND FLOWERS, P. Good in parts: The Gay Men's Task Force in Glasgow--A response to Kelly. AIDS Care 16:15-165,2004. PMID: 14676022

HONG, F.C.; ZHOU, H.; CAI, Y.M.; ET AL. Prevalence of syphilis and HIV infections among men who have sex with men from different settings in Shenzhen, China: Implications for HIV/STD surveillance. Sexually Transmitted Infections 85:42-44, 2009. PMID: 18653567

KALICHMAN, S.C.; CHERRY, C.; AND BROWNE-SPERLING, F. Effectiveness of a video-based motivational skills-building HIV risk-reduction intervention for inner-city African American men. Journal of Consulting and Clinical Psychology 67:959-966, 1999. PMID: 10596517

KALICHMAN, S.C.; SIMBAYI, L.C.; VERMAAK, R.; ET AL. HIV/AIDS risks among men and women who drink at informal alcohol serving establishments (Shebeens) in Cape Town, South Africa. Prevention Science 9:55-62,2008. PMID: 18264762

KALICHMAN, S.C.; SIMBAYI, L.C.; VERMAAK, R. ET AL. HIV/AIDS risk reduction counseling for alcohol using sexually transmitted infections clinic patients in Cape Town, South Africa. Journal of Acquired Immune Deficiency Syndromes 44:594-600, 2007. PMID: 17325606

KAPIGA, S.H.; SAM, N.E.; SHAO, J.F.; ET AL. Herpes simplex virus type 2 infection among bar and hotel workers in northern Tanzania: Prevalence and risk factors. Sexually Transmitted Diseases 30:187-192, 2003. PMID: 12616132

KAPIGA, S.H.; SAM, N.E.; SHAO, J.F.; ET AL. HIV-1 epidemic among female bar and hotel workers in northern Tanzania: Risk factors and opportunities for prevention. Journal of Acquired Immune Deficiency Syndromes 29:409-417, 2002. PMID: 11917247

KEGELES, S.M.; HAYS, R.B.; AND COATES, T.J. The Mpowerment Project: A community-level HIV prevention intervention for young gay men. American Journal of Public Health 86:1129-1136, 1996. PMID: 8712273

KELLY, J.A.; AMIRKHANIAN, Y.A.; KABAKCHIEVA, E.; ET AL. Prevention of HIV and sexually transmitted diseases in high risk social networks of young Roma (Gypsy) men in Bulgaria: Randomised controlled trial. British Medical Journal 333(7578): 1098, 2006. PMID: 17040924

KELLY, J.A. Popular opinion leaders and HIV prevention peer education: Resolving discrepant findings, and implications for the development of effective community programmes. AIDS Care 16:139150, 2004. PMID: 14676020

KELLY, J.A.; ST. LAWRENCE, J.S.; DIAZ, Y.E.; ET AL. HIV risk behavior reduction following intervention with key opinion leaders of a population: An experimental community-level analysis. American Journal of Public Health 81:168-171,1991. PMID: 1990853

KELLY, J.A.; ST. LAWRENCE, J.S.; STEVENSON, L.Y.; ET AL. Community AIDS/HIV risk reduction: The effects of endorsements by popular people in three cities. American Journal of Public Health 82:14831489, 1992. PMID: 1443297

KELLY, J.A.; MURPHY, DA.; SIKKEMA, K.J.; ET AL. Randomized, controlled, community-level HIV prevention intervention for sexual-risk behaviour among homosexual men in US cities. Lancet 350:1500-1505,1997. PMID: 9388397

KHAN, M.R.; RASOLOFOMANANA, J.R.; McCLAMROCH, K.J.; ETAL. High-risk sexual behavior at social venues in Madagascar. Sexually Transmitted Diseases 35:738-745,2008. PMID: 18496471

KINCAID, D.L. From innovation to social norm: Bounded normative influence. Journal of Health Communication 9 (Suppl. 1): 37-57, 2004. PMID: 14960403

LATKIN, C.; HUA, W.; AND DAVEY, M.A. Factors associated with peer HIV prevention in drug-using communities. AIDS Education and Prevention 16:499-508, 2004. PMID: 15585427

LEWIS, J.J.; GARNETT, G.P.; MHLANGA, S.; ET AL. Beer halls as a focus for HIV prevention activities in rural Zimbabwe. Sexually Transmitted Diseases 32:364-369, 2005. PMID: 15912083

MAH, T.L., AND HALPERIN, D.T. Concurrent sexual partnerships and the HIV epidemics in Africa: Evidence to move forward. AIDS and Behavior . [Epub ahead of print] PMID: 18648926

MATAURE, P.; MCFARLAND, W.; FRITZ, K.; ET AL. Alcohol use and high-risk sexual behavior among adolescents and young adults in Harare, Zimbabwe. AIDS and Behavior 6:211-219, 2002.

MBULAITEYE, S.M.; RUBERANTWARI, A.; NAKIYINGI, J.S.; ET AL. Alcohol and HIV: A study among sexually active adults in rural southwest Uganda. International Journal of Epidemiology 29:911-915,2000. PMID: 11034977

MILLER, R.L.; KLOTS, D.; AND ECKHOLDT, H.M. HIV prevention with male prostitutes and patrons of hustler bars: Replication of an HIV prevention intervention. American Journal of Community Psychology 26:97-131, 1998. PMID: 9574500

MORISKY, D.E.; STEIN, J.A.; CHIAO, C.; ET AL. Impact of a social influence intervention on condom use and sexually transmitted infections among establishment-based female sex workers in the Philippines: A multilevel analysis. Health Psychology 25:595-603,2006. PMID: 17014277

MOROJELE, N.K.; KACHIENG'A, M.A.; MOKOKO, E.; ET AL. Alcohol use and sexual behaviour among risky drinkers and bar and shebeen patrons in Gauteng province, South Africa. Social Science and Medicine 62:217-227, 2006. PMID: 16054281

NIMH Collaborative HIV/STD Prevention Trial. "Effectiveness of a Community-Level HIV/STD Prevention Intervention in Promoting Safer Sexual Behaviors in High-Risk Populations." Presented at the International AIDS Conference, Mexico City, 2008.

NARDONE, A.; FRANKIS, J.S.; DODDS, J.P.; ET AL. A comparison of high-risk sexual behaviour and HIV testing amongst a bar-going sample of homosexual men in London and Edinburgh. European Journal of Public Health 11:185-189, 2001. PMID: 11420808

PALEPU, A.; RAJ, A.; HORTON, N.J.; ET AL. Substance abuse treatment and risk behaviors among HIV-infected persons with alcohol problems. Journal of Substance Abuse Treatment 25:3-9, 2005. PMID: 15723726

ROGERS, E. Diffusion of Innovations, 5th ed. New York: Free Press, 2003.

ROU, K.; WU, Z.; SULLIVAN, S.G.; ET AL. A five-city trial of a behavioural intervention to reduce sexually transmitted disease/HIV risk among sex workers in China. AIDS 21(Suppl. 8): S95-S101, 2007. PMID: 18172399

SHISANA, O.; REHLE, T.; SIMBAYI, L.; ET AL. South African National HIV Prevalence, Incidence, Behaviour and Communication Survey 2005. Cape Town: Human Sciences Research Council Press, 2005.

SIMON ROSSER, B.; WEST, W.; AND WEINMEYER, R. Are gay communities dying or just in transition; Results from an international consultation examining possible structural change in gay communities. AIDS Care 5:588-595,2008. PMID: 18484330

SINGH, K.; SAMBISA, W.; MUNYATI, S.; ET AL. Targeting HIV interventions for adolescent girls and young women in Southern Africa: Use of the PLACE Methodology in Hwange District, Zimbabwe. AIDS and Behavior, 2009 [Epub ahead of print]. PMID: 19452272

VARAM, S.; SRIKISHNAN, A.K.; LATKIN, C.; ET AL. Male alcohol use and unprotected sex with non regular partners: Evidence from wine shops in Chennai, India. Drug and Alcohol Dependence 94:133-141,2008. PMID: 18187270

WEINHARDT, L.S., AND CAREY, M.P. Does alcohol lead to sexual risk behavior? Findings from event-level research. Annual Review of Sex Research 11:125-157,2000. PMID: 11351830

WEIR, S.S.; FIGUEROA, J.P.; BYFIELD, L.; ET AL. Randomized controlled trial to investigate impact of site-based safer sex programmes in Kingston, Jamaica: Trial design, methods, and baseline findings. Tropical Medicine and International Health 13:801-813, 2008. PMID: 18482079

WEIR, S.S.; PAILMAN, C.; MAHLALELA, X.; ET AL. From people to places: Focusing AIDS prevention efforts where it matters most. AIDS 17:895-903, 2003. PMID: 12660537

XIA, Q.; THOLANDI, M.; OSMOND, D.H.; ET AL. The effect of venue sampling on estimates of HIV prevalence and sexual risk behaviors in men who have sex with men. Sexually Transmitted Diseases 33:545-550, 2006. PMID: 16735957

ZACHARIAH, R.; SPIELMANN, M.P.; HARRIES, A.D.; ET AL. Sexually transmitted infections and sexual behaviour among commercial sex workers in a rural district in Malawi. International Journal of STD & AIDS 14:185-188, 2003. PMID: 12665441

Seth C. Kalichman, Ph.D.

SETH C. KALICHMAN, PH.D., is professor of psychology at the University of Connecticut, Storrs, Connecticut.
Table 1 Selected International Studies of HIV Risks in
Alcohol-Serving Venues

Authors         Samples           Summary of Findings

Singh 2009      511 young         Young women who socialize at
                women, 15-24      drinking places demonstrated the
                years old,        greatest risk for STI/HIV, with 35
                Zimbabwe          percent reporting two or more sex
                                  partners in the previous year
                                  (compared with 5 percent of women
                                  in other venues) and 30 percent
                                  reporting exchanging sex for money
                                  in the past year (compared with 10
                                  percent of women in nonalcohol

Hong            1,146 men in      HIV and syphilis prevalence were
et al. 2009     gay saunas,       higher in gay saunas than gay
                bars, and         bars. Demographic characteristics
Sivaram         clinics, China    also differed substantially across
et al. 2008     1,196 male wine   venues with younger,
                shop/bar          gay-identified men attending bars
                patrons, India    and older married men attending
                                  saunas. Nearly one in four people
                                  had four or more sex partners, 89
                                  percent used alcohol in
                                  conjunction with sex, and alcohol
                                  use was associated with
                                  unprotected sex.

Kalichman       91 men and 248    Participants who meet sex partners
et al. 2008     women who drink   in the drinking establishment (28
                in informal       percent) had higher risk for HIV
                bars              infection compared with patrons
                (shebeens),       who did not meet sex partners in
                South Africa      shebeens. Men and women who meet
                                  sex partners in shebeens did not
                                  differ in their risk behaviors.

Grov            886 men who       Exact venue where sex partners
et al. 2007     have sex with     meet, including bars, bath houses,
                men recruited     private sex parties, gyms, public
                at venues where   areas, and the Internet did not
                they meet sex     relate to patterns of HIV risk
                partners, New     behavior. Personal characteristics
                York and Los      and identity were related to risk
                Angeles           practices across venues.

Xia             398 men who       Three of four men visited a gay
et al. 2006     have sex with     bar at least once in the previous
                men and meet      year. Men who go to bars
                sex partners in   demonstrate high rates of sexual
                various venues,   risk behavior compared with men
                California        who do not go to any gay venues
                                  and less risk than men who attend
                                  circuit parties

Zachariah       1,817 female      Sex workers with symptomatic STIs
et al. 2003     sex workers,      who drank were four times more
                Malawi            likely to not use condoms than
                                  were STI symptomatic women who did
                                  not drink (37 percent vs. 12

Weir            3,085 men,        People in townships and business
et al. 2003     1,564 women in    districts meet sex partners in
                alcohol venues    drinking establish- ments and
                in town- ships    sexual risk is closely associated
Kapiga          and business      with venues where drinking occurs.
et al. 2003     districts,        Alcohol use by men once a week was
                South Africa      associated with increased risk of
                309 female and    STIs compared with nondrinkers,
                206 male bar      particularly risks for contracting
                and hotel         herpes simplex virus.

Kapiga et       312 women         Women who worked in bars
al. 2002        working in bars   demonstrated 26 percent HIV
                and hotels,       prevalence, with risk for HIV 2.5
                Tanzania          times greater for women who drank
                                  2 or more days per week.

Fritz           324 men           HIV prevalence increased with
et al. 2002     attending beer    greater frequency of drinking.
                halls, Zimbabwe   Eleven percent of men reporting no
                                  alcohol use were HIV positive as
                                  were 41 percent of men that drank
                                  on 15 or more of the last 30 days.
                                  HIV prevalence and incidence were
                                  related to meeting sex partners in
                                  beer halls and having sex while

Mataure         227 men and 78    Sixty percent of young women at
et al. 2002     women, 15 to 21   drinking establishments usually
                years old, at     meet sex partners there. Forty-two
                nightclubs and    percent had sex after drinking in
                bottle stores,    the past 90 days.

Nardone         2,397 men who     One-third of men practiced recent
et al. 2002     have sex with     unprotected anal intercourse with
                men from gay      more than half of men having been
                bars, London      tested for HIV. Differences were
                and Edinburgh     observed between cities, with
                                  higher risk apparent in London.

NOTE: HIV, human immunodeficiency virus; STI, sexually transmitted

Table 2 HIV Risk Reduction Interventions Based in Alcohol Venues in
International Settings

                  Target            Intervention
                  Population        and Study         Summary of
Authors           and Venues        Design            Findings

Social Influence Interventions

NIMH              Wine shop         C-POL             Comparable
Collaborative     customers in      intervention;     reductions in
HIV/STD           impoverished      24 wine shops     HIV/STI
Prevention        slums, India      within            transmission
Trial 2008                          impoverished      risk behaviors
                                    slum communi-     occurred in the
                                    ties;             intervention
                                    community-level   and control
                                    RCT conducted     communities.
                                    in India and      Both conditions
                                    four other        also
                                    countries;        demonstrated
                                    observation       comparable
                                    over 2 years.     reductions in
                                                      new STI.

Miller            Gay bars with     Adaptation of     The
1998              significant       POL               intervention
                  male              intervention;     demonstrated
                  prostitution in   multiple          significant
                  New York City     baseline quasi-   reductions in
                                    experimental      paid USI across
                                    design with       venues.
                                    staggered         Frequencies of
                                    intervention      unpaid USI were
                                    implementation;   not affected by
                                    observations      the
                                    occurred in       intervention.
                                    each bar five     Bars with more
                                    times at          stable patrons
                                    2-month           and Latino and
                                    intervals.        White men
                                                      benefited from
                                                      whereas Black
                                                      men did not.

Kelly             Gay bars in       POL               Intervention
et al. 1997       eight small       intervention,     cities
                  U.S. cities       community-level   demonstrated
                                    RCT;              less USI than
                                    observation       control cities
                                    over 1-year       by a factor of
                                    postintervention. nearly four-
                                                      fold. Condom
                                                      use increased
                                                      from 45 percent
                                                      protected acts
                                                      to 67 percent,
                                                      with no
                                                      occurring in
                                                      the control

Kelly             Gay bars in       POL               The POL model
et al. 1992       three             intervention;     intervention
                  southeastern      multiple          demonstrated 15
                  U.S. small        baseline quasi-   percent to 24
                  cities            experimental      percent
                                    design with       reductions in
                                    staggered         USI across all
                                    intervention      three cities.
                                    implementation;   Similar
                                    observations      outcomes
                                    between 3 and 9   occurred for
                                    months            numbers of sex
                                    postintervention. partners and
                                                      increases in
                                                      condom use
                                                      during anal

Structural Interventions

Rou               Female sex        Enhanced STI      Reductions in
et al. 2008       workers in        clinic services   multiple risk
                  bars, massage     with STI/HIV      indicators were
                  parlors, dance    education and     observed across
                  halls, and        sex worker        venues during
                  beau- ty          condom            the observa-
                  parlors, China    promotion         tion period.
                                    outreach; five    Condom use with
                                    sites in two      last client
                                    provinces         increased from
                                    participated in   55 percent to
                                    the public        67 percent,
                                    health            prevalence of
                                    evaluation with   gonorrhea
                                    no control        declined from
                                    condition;        26 percent to 4
                                    1-year obser-     percent, and
                                    vation period.    prevalence of
                                                      decreased from
                                                      41 percent to
                                                      26 percent.

Blank             Gay bars and      "Hot Shot"        Implementation
et al. 2005       clubs             public health     of portable
                  identified        program           health services
                  through           offering          in venues
                  syphilis case     screening,        resulted in
                  inter- views,     diagnostic, and   four new
                  New York City     referral          syphilis and
                                    services to men   seven new HIV
                                    in venues         detections.
                                    during evening    Packaging STI
                                    hours;            and HIV
                                    uncontrolled      services with
                                    public health     general health
                                    evaluation;       programming was
                                    1-year            deemed feasible
                                    observation       and potentially
                                    period.           effective.

Multilevel Social Influence and Structural Interventions

Morisky           Female sex        Four              Increases in
et al. 2006       workers in        conditions:       condom use at
                  bars,             peer counseling   last sex and
                  nightclubs,       and social        reductions in
                  discos, beer      influence for     self-reported
                  gardens, and      HIV risk          STI occurred in
                  karaoke bars,     reduction,        the combined
                  Philippines       manager           peer/manager
                                    training to       intervention
                                    rein-force        condition
                                    employee health   compared with
                                    and structural    the control
                                    health            site. These
                                    improvement,      effects were
                                    peer counseling   bolstered by
                                    with manager      more positive
                                    training,         condom
                                    no-intervention   attitudes and
                                    control           increased
                                    condition;        condom
                                    quasi-            promotion
                                    experimental      within the
                                    design with       venues. Sites
                                    venues nested     receiving
                                    in four cities    either the peer
                                    randomized to     or manager
                                    one of four       interventions
                                    conditions;       demonstrated
                                    2-year            variable
                                    observation       patterns of
                                    period.           risk reduction
                                                      relative to the
                                                      control site.

Flowers           Gay bars in       Peer-led sexual   Peer education
et al. 2002       Glasgow,          health            and associated
                  Scotland          promotion         services failed
                                    conducted in      to demonstrate
                                    venues,           reductions in
                                    gay-specific      USI and
                                    sexual health     increases in
                                    services in       knowledge of
                                    hospitals, and    sex partner HIV
                                    free              status or
                                    phone-hotline     negotiated
                                    service; two-     safety.
                                    city quasi-       Increases in
                                    experimental      HIV testing and
                                    trial; 3-year     hepatitis B
                                    observation       vaccination
                                    period.           were observed
                                                      for men with
                                                      direct contact
                                                      with the
                                                      but not for the
                                                      community in

NOTE: C-POL, community popular opinion leader; HIV, human
immunodeficiency virus; POL, popular opinion leader; RCT,
randomized community trial; STI, sexually transmitted
infection; USI, unprotected sexual intercourse.
Gale Copyright: Copyright 2010 Gale, Cengage Learning. All rights reserved.